total_retry_time is not "time since last connect"

• Version: 2.8.0
• Platform: NodeJS 11.12 on Windows 10
• Description:
  total_retry_time is not "time since last connect", it's just total time of waiting for retry, time from connect to timeout was ignored.

Here is my log with real time and time passed to retry_strategy:

{"name":"App","hostname":"Desktop","pid":28432,"level":50,"msg":"[RedisClient] Redis losted connection, time since last connect: 0 { attempt: 1,\n  error:\n   { Error: Redis connection to redis:6379 failed - connect ETIMEDOUT 192.168.31.25:6379\n       at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1083:14)\n     errno: 'ETIMEDOUT',\n     code: 'ETIMEDOUT',\n     syscall: 'connect',\n     address: '192.168.31.25',\n     port: 6379 },\n  total_retry_time: 0,\n  times_connected: 0 }","time":"2019-03-18T08:25:38.402Z","v":0}
{"name":"App","hostname":"Desktop","pid":28432,"level":50,"msg":"[RedisClient] Redis losted connection, time since last connect: 10 { attempt: 2,\n  error:\n   { Error: Redis connection to redis:6379 failed - connect ETIMEDOUT 192.168.31.25:6379\n       at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1083:14)\n     errno: 'ETIMEDOUT',\n     code: 'ETIMEDOUT',\n     syscall: 'connect',\n     address: '192.168.31.25',\n     port: 6379 },\n  total_retry_time: 10,\n  times_connected: 0 }","time":"2019-03-18T08:25:59.424Z","v":0}
{"name":"App","hostname":"Desktop","pid":28432,"level":50,"msg":"[RedisClient] Redis losted connection, time since last connect: 20 { attempt: 3,\n  error:\n   { Error: Redis connection to redis:6379 failed - connect ETIMEDOUT 192.168.31.25:6379\n       at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1083:14)\n     errno: 'ETIMEDOUT',\n     code: 'ETIMEDOUT',\n     syscall: 'connect',\n     address: '192.168.31.25',\n     port: 6379 },\n  total_retry_time: 20,\n  times_connected: 0 }","time":"2019-03-18T08:26:20.448Z","v":0}
{"name":"App","hostname":"Desktop","pid":28432,"level":50,"msg":"[RedisClient] Redis losted connection, time since last connect: 30 { attempt: 4,\n  error:\n   { Error: Redis connection to redis:6379 failed - connect ETIMEDOUT 192.168.31.25:6379\n       at TCPConnectWrap.afterConnect [as oncomplete] (net.js:1083:14)\n     errno: 'ETIMEDOUT',\n     code: 'ETIMEDOUT',\n     syscall: 'connect',\n     address: '192.168.31.25',\n     port: 6379 },\n  total_retry_time: 30,\n  times_connected: 0 }","time":"2019-03-18T08:26:41.469Z","v":0}

The documents says it's milliseconds since last connect, but when the callback called with "total_retry_time: 30", 1 minutes was passed...

Start at the retry_strategy callback and trace how total_retry_time is calculated during the connection timeout shown in the report. Reproduce the logged timing sequence on NodeJS, then verify that the reported value includes the elapsed connect-to-timeout period as well as retry waits.